The experience of one person, or even a handful, do not in any way negate an 
obvious and growing trend in the software industry--directly related to "agile" 
development--to consider TW involvement as pointless until the final iteration. 

Yes, there are organizations that still do business as they did 20 or 30 years 
ago, just as there are still organizations using COBOL, SNOBOL, and other odd 
applications. If their system works, more power to them, and to the TWs they 

The difference is in whether or not the organization is developing software, or 
creating an application that "implements the vision" of a handful of movers and 
shakers at the top. That handful can do as they please, whether or not it is of 
long-term benefit to the organization. For software developed in a competitive 
marketplace, the role of the TW is rapidly changing to a diminished involvement. in the Design, Development, and 
Production of:Technical Documentation - Online Content - Enterprise Websites

From: lhs_emf at pacbell.netTo: tekwrytr at; 
framers@lists.frameusers.comSubject: RE: radical revamping of techpubsDate: 
Mon, 29 Oct 2007 21:14:18 -0500

This may be your experience, in my experience in fact there is no IF about it, 
I just put it that way to be gentile.

Our documents pre-sage multi-mullion dollar contracts (at each stage of the 
project) and there is always plenty of fuzzy concepts to go around at the early 
stages. No documents, no contracts. 

TWs and in particular the directors, managers are involved at these stages. 
Documentation is a 100% necessary adjunct to business development from the 

From: Technical Writer [mailto:tekwrytr at] Sent: Monday, October 
29, 2007 8:29 PMTo: Leslie H Schwartz; framers@lists.frameusers.comSubject: RE: 
radical revamping of techpubs

That is a very big if. A full partner participant-stakeholder, or more likely 
the department manager? It is more likely that the software developers, 
business analysts, and the project manager are collaborating to get a decent 
set of requirements down. At that stage, TWs have no place, whether department 
managers, full partner participant-stakeholders, or something else. When the 
requirements are determined, and possibly after several iterations, possibly 
after a prototype is up and running, TWs might be brought in. Even at that 
stage, it is early, because the GUI crew may not have the interface coded, the 
developers might not have the functionality carved in stone, and everything is 
still uncertain (in regards to exactly what the final product will be and do). 
TWs complete a very necessary task; creating user assistance. Until the final 
iteration, until all the requirements have been met, until there is little or 
no possibility of changes to the end product, there is little point in 
generating documentation that might become obsolete at the next iteration. in the Design, Development, and 
Production of:Technical Documentation - Online Content - Enterprise Websites

Date: Mon, 29 Oct 2007 08:26:46 -0700From: lhs_emf at pacbell.netSubject: Re: 
radical revamping of techpubsTo: tekwrytr at hotmail.comCC: framers at

Actually, I disagee, if the TW is a full partner participant - stakeholder, or 
more likely the department manager in the scenario you are discussing, they 
should also participate early on to get the sense of the uncertainty and what 
those issues are, at the very least these issues are going to affect their 
scheduling and the expectations they have to deal with.

----- Original Message ----From: Technical Writer <tekwrytr at>To: 
Leslie Schwartz <lhs_emf at>; framers at lists.frameusers.comSent: 
Monday, October 29, 2007 8:44:16 AMSubject: RE: radical revamping of techpubsI 
agree wholeheartedly. That is not the issue. The issue goes back to the BA 
interpretation of (and translation of) the software requirements. If there is a 
high level of certainty on the client side about what the finished product 
should be, TWs should start early. If not, and it is essentially a fishing 
expedition with ambiguous outcome, TWs are only useful at the last. 
Unfortunately, the "agile" methodologies strongly sell the sense of control to 
executives, pushing the idea that they can develop on the fly, adding and 
removing "requirements" as the executives see fit. in the Design, Development, and 
Production of:Technical Documentation - Online Content - Enterprise Websites> 
From: lhs_emf at> To: tekwrytr at; bhechter at; framers at> Subject: RE: radical revamping 
of techpubs> Date: Sun, 28 Oct 2007 19:04:10 -0500> > I belong to several 
message - interest groups and I am used to hearing people give their opinions 
in a bombastic manner. So its no> big deal to see that happening here. But if 
this discussion is to have any real value it will be to share our perspectives 
with> others and learn something about points of view's entirely different than 
our own, which requires some tolerance and mutual respect.> > > My view and 
experience is that it definitely helps to get the TW involved early on, but 
it?s a waste of time for them to sit all the> way through each meeting, and for 
the entire duration of each meeting.> > Marketing requirements documents and 
engineering specification documents, if they are adequately written will help 
the TW formulate> the user documentation at a fairly early stage, but the bulk 
of the documentation effort comes towards the end of the development> cycle. 
And ideally the writer of the user guide if that is they type of documentation 
we are discussing now, should be a> knowledgeable user with some fresh insights 
into the learning curve the novice user will face, and some empathy for that 
new user.> > Ignoring the need for documentation, putting it off until the last 
moment is a formula for poor quality documentation.> > - In my humble opinion.> 
> Have a great work week!> > Leslie> > > -----Original Message-----> From: at 
[ at] On> Behalf 
Of Technical Writer> Sent: Sunday, October 28, 2007 5:47 PM> To: bhechter at; framers at> Subject: RE: radical revamping 
of techpubs> > > Well, a difference of opinion is what makes a horse race. 
Iterative software methods do not require iterative documentation methods;> in 
most cases, documentation before the last iteration is considered both wasteful 
and useless. While I have a great deal of respect> for Steve McConnell, 
proposing early draft user guides as a replacement for requirement specs is a 
bit off the road. > > If you develop software, and intend to use early draft 
user guides instead of requirements, you are going to be greeting the folks> at 
Wal-Mart rather than trying to pull back a contract or two from Bangalore. The 
statement is at odds with most developers' (and> most business analysts') 
understanding of "requirements." Putting an occasional "agile" into a sentence 
doesn't make the process any> more reasonable. > > I didn't invent the idea of 
ignoring documentation until the final product is ready (or almost ready) to 
ship. Far more intelligent,> competent, and capable people than me have decided 
that "involving TWs from the early stages of development" is only useful when 
the> end product is carved in stone before the first line of code is written. 
That, for better of worse, is rarely the case.> > Lastly, given that about a 
third of all software projects, agile or otherwise, fail so badly they are 
abandoned, if you ignore> documentation completely, you have a one in three 
chance of coming out ahead when the project flops because you have at least 
saved> the cost of documentation.> in the 
Design, Development, and Production of:Technical Documentation - Online Content 
-> Enterprise Websites> > > Date: Sun, 28 Oct 2007 12:21:17 -0700From: bhechter 
at yahoo.comSubject: re: radical revamping of techpubsTo: tekwrytr at 
hotmail.comCC:> framers at lists.frameusers.comSorry, but I find the thread 
both:a) Off-topicb) Misleading. Iterative sofware methods require iterative> 
documentation methods, but by no means do they eliminate the parallel need for 
early draft user manuals. In fact, Steve McConnell> (Code Complete) proposes 
early draft user guides as an agile replacement for requirements specs.Ben> 
Because the application itself> is built in an iterative process, rather than > 
being carved in stone, reacting to feedback from the client, documentation > 
before> the last minute is pointless. The reason should be obvious; the > 
application being documented in the early stages bears little> resemblance > to 
the application delivered. Ben Hechter Vancouver BC bhechter at> 
_________________________________________________________________> Windows Live 
Hotmail and Microsoft Office Outlook ? together at last.  Get it now.>>
 > > You are currently subscribed to Framers as lhs_emf at> > Send 
list messages to framers at> > To unsubscribe send a 
blank email to > framers-unsubscribe at> or visit> > 
Send administrative questions to listadmin at Visit> for more resources and info.> Boo! Scare away worms, 
viruses and so much more! Try Windows Live OneCare! Try now! 

Boo! Scare away worms, viruses and so much more! Try Windows Live OneCare! Try 
Boo!?Scare away worms, viruses and so much more! Try Windows Live OneCare!

Reply via email to